projects / oota-llvm.git / history
? search:
summary | shortlog | log | commit | commitdiff | tree
first ⋅ prev ⋅ next
AMDGPU: Fix crash with dispatch.ptr intrinsic with non-HSA target
[oota-llvm.git] / lib / Target / AMDGPU / SIISelLowering.cpp
2016-01-11 Matt ArsenaultAMDGPU: Fix crash with dispatch.ptr intrinsic with... blob | commitdiff | raw
2016-01-11 Matt ArsenaultAMDGPU: Pattern match ffbh pattern to instruction. blob | commitdiff | raw | diff to current
2016-01-11 Matt ArsenaultAMDGPU: Remove dead target dag combine blob | commitdiff | raw | diff to current
2015-12-15 Tom StellardAMDGPU/SI: Select constant loads with non-uniform addre... blob | commitdiff | raw | diff to current
2015-12-15 Tom StellardAMDGPU/SI: Add llvm.amdgcn.v.interp.p[12] intrinsics blob | commitdiff | raw | diff to current
2015-12-14 Matt ArsenaultAMDGPU: Use generic bitreverse intrinsic blob | commitdiff | raw | diff to current
2015-12-10 Tom StellardAMDGPU/SI: Emit constant arrays in the .text section blob | commitdiff | raw | diff to current
2015-12-10 Tom StellardAMDGPU/SI: Add support for sgpr and vgpr inline assembl... blob | commitdiff | raw | diff to current
2015-12-01 Matt ArsenaultAMDGPU: Implement isNoopAddrSpaceCast blob | commitdiff | raw | diff to current
2015-12-01 Tom StellardAMDGPU/SI: Remove REGISTER_STORE/REGISTER_LOAD code... blob | commitdiff | raw | diff to current
2015-11-30 Matt ArsenaultAMDGPU: Fix unused function blob | commitdiff | raw | diff to current
2015-11-30 Matt ArsenaultAMDGPU: Rework how private buffer passed for HSA blob | commitdiff | raw | diff to current
2015-11-30 Matt ArsenaultAMDGPU: Rename enums to be consistent with HSA code... blob | commitdiff | raw | diff to current
2015-11-30 Matt ArsenaultAMDGPU: Remove SIPrepareScratchRegs blob | commitdiff | raw | diff to current
2015-11-30 Matt ArsenaultAMDGPU: Use assert zext for workgroup sizes blob | commitdiff | raw | diff to current
2015-11-30 Matt ArsenaultAMDGPU: Don't reserve SCRATCH_PTR input register blob | commitdiff | raw | diff to current
2015-11-26 Tom StellardAMDGPU: Add llvm.amdgcn.dispatch.ptr intrinsic blob | commitdiff | raw | diff to current
2015-11-25 Matt ArsenaultAMDGPU: Make v2i64/v2f64 legal types. blob | commitdiff | raw | diff to current
2015-11-24 Matt ArsenaultAMDGPU: Split LDS vector loads blob | commitdiff | raw | diff to current
2015-11-24 Matt ArsenaultAMDGPU: Split x8 and x16 vector loads instead of scalarize blob | commitdiff | raw | diff to current
2015-11-02 Matt ArsenaultAMDGPU: Error on graphics shaders with HSA blob | commitdiff | raw | diff to current
2015-10-29 Marek OlsakAMDGPU/SI: handle undef for llvm.SI.packf16 blob | commitdiff | raw | diff to current
2015-10-21 Matt ArsenaultAMDGPU: Simplify VOP3 operand legalization. blob | commitdiff | raw | diff to current
2015-10-20 Matt ArsenaultAMDGPU: Add MachineInstr overloads for instruction... blob | commitdiff | raw | diff to current
2015-10-06 Tom StellardAMDGPU/SI: Remove calling convention assertion from... blob | commitdiff | raw | diff to current
2015-09-29 Marek OlsakAMDGPU/SI: Don't set DATA_FORMAT if ADD_TID_ENABLE... blob | commitdiff | raw | diff to current
2015-09-25 Matt ArsenaultAMDGPU: Re-justify workaround and fix worked around... blob | commitdiff | raw | diff to current
2015-09-22 NAKAMURA TakumiReformat comment lines. blob | commitdiff | raw | diff to current
2015-09-21 Craig TopperUse makeArrayRef or None to avoid unnecessarily mention... blob | commitdiff | raw | diff to current
2015-09-16 Sanjay Patelpropagate fast-math-flags on DAG nodes blob | commitdiff | raw | diff to current
2015-09-09 Matt ArsenaultSelectionDAG: Support Expand of f16 extloads blob | commitdiff | raw | diff to current
2015-09-03 Sanjay Patelcheck for fastness before merging in DAGCombiner::Merge... blob | commitdiff | raw | diff to current
2015-08-08 Benjamin KramerFix some comment typos. blob | commitdiff | raw | diff to current
2015-08-07 Matt ArsenaultAMDGPU: Assume SMRD access for constant address space blob | commitdiff | raw | diff to current
2015-08-01 Craig TopperDe-constify pointers to Type since they can't be modifi... blob | commitdiff | raw | diff to current
2015-07-31 Matt ArsenaultAMDGPU: Fix v16i32 to v16i8 truncstore blob | commitdiff | raw | diff to current
2015-07-20 Tom StellardAMDGPU/SI: Add VI patterns to select FLAT instructions... blob | commitdiff | raw | diff to current
2015-07-16 Tom StellardAMDPGU/SI: Use AssertZext node to mask high bit for... blob | commitdiff | raw | diff to current
2015-07-10 Matt ArsenaultAMDGPU: Fix chains for memory ops dependent on argument... blob | commitdiff | raw | diff to current
2015-07-10 Matt ArsenaultAMDGPU: Use requested chain when lowering arguments blob | commitdiff | raw | diff to current
2015-07-09 Tom StellardAMDGPU: Add helper function for implicit parameter... blob | commitdiff | raw | diff to current
2015-07-09 Mehdi AminiRe-instate the EVT parameter to getScalarShiftAmountTy... blob | commitdiff | raw | diff to current
2015-07-09 Mehdi AminiRemove getDataLayout() from TargetLowering blob | commitdiff | raw | diff to current
2015-07-09 Mehdi AminiMake isLegalAddressingMode() taking DataLayout as an... blob | commitdiff | raw | diff to current
2015-07-09 Mehdi AminiMake TargetLowering::getShiftAmountTy() taking DataLayo... blob | commitdiff | raw | diff to current
2015-07-09 Mehdi AminiMake TargetLowering::getPointerTy() taking DataLayout... blob | commitdiff | raw | diff to current
2015-07-05 Benjamin Kramer[TargetLowering] StringRefize asm constraint getters. blob | commitdiff | raw | diff to current
2015-06-26 Tom StellardAMDGPU/SI: There are no implicit kernel args in the... blob | commitdiff | raw | diff to current
2015-06-23 Matt ArsenaultAMDGPU: Use getAsInteger instead of atoi blob | commitdiff | raw | diff to current
2015-06-13 Tom StellardR600 -> AMDGPU rename blob | commitdiff | raw | diff to current
2012-07-16 Tom StellardRevert "AMDGPU: Add core backend files for R600/SI... blob | commitdiff | raw
2012-07-16 Tom StellardAMDGPU: Add core backend files for R600/SI codegen v6 blob | commitdiff | raw | diff to current
C/C++ LLVM-based compilers that forbids OOTA behaviors